Ehsan Yadegari; Akbar Alem Tabriz; Mostafa Zandieh
Abstract
Over the past decade, due to environmental laws and the competitive environment, development of an effective tactical plan for efficient and integrated supply chain and considering the responsibility of organizations to collect defective goods seems impossible. In this paper a mixed-integer linear programming ...
Read More
Over the past decade, due to environmental laws and the competitive environment, development of an effective tactical plan for efficient and integrated supply chain and considering the responsibility of organizations to collect defective goods seems impossible. In this paper a mixed-integer linear programming is considered to mathematically model the essentially five stages along our supply chain network: suppliers, manufacturers, DCs, customers, and Dismantlers.Delivers raw materials from suppliers to factories and then through distribution centers, delivering the final product to customers. On the other hand, it simultaneously collects recycled goods from customers and enters the cycle of safe reconstruction or destruction.The aim of this model is minimizing the costs of establishing facilities at potential points as well as the optimal flow of materials in the network layers. Since the problem is NP-hard, to solve it, the cloud theory based simulated annealing algorithm has been used. We also used the tree-covering method to show the answer, which uses fewer arrays than other methods in the literature. To analyze the accuracy and speed of the proposed algorithm, we compared its performance with the genetic and simulated annealing algorithm. The results show that the cost function in the cloud-based refrigeration simulation algorithm provides more accurate answers than both algorithms studied in the literature. The results show that the cost function in the cloud-based simulated annealing algorithm provides more accurate answers than both algorithms studied in the literature. Also, in terms of convergence rate criterion, the proposed method has better position than the genetic algorithm, but it is not significantly different from simulated annealing algorithm.
